TRIBOLOGY LETTERS, cilt.61, 2016 (SCI İndekslerine Giren Dergi)
Understanding soot nanoparticle interaction with oil additives and the causes of soot-induced thickening would assist in lubricant formulation, prolonging engine life and improving engine efficiency. Three-dimensional measurement of soot structures is currently not undertaken as established techniques are limited to two dimensions. While they give valuable information on the structure and reactivity of soot nanoparticles, it is not easy to correlate this to geometry of primary particles and agglomerates. In this work, we investigate the development and application of 3D-TEM for characterisation of soot agglomerates as a new capability to yield information on the volumetric character of fractal nanoparticles. This investigation looks at the feasibility for volume reconstruction of nanometric soot particles in used engine oil from multiple imaging at different tilt angles. Bright-field TEM was used to capture two-dimensional images of soot. Heptane and diethyl ether washes were used to remove volatile contaminants and allowed for images from -60 degrees to +60 degrees tilt with no sign of carbon build-up to be acquired. Tomographic reconstruction from the aligned tilt-series images based on weighted back-projection algorithm has yielded useful information about complex soot nanoparticle size. Estimation of soot mass in oil by nanoparticle tracking analysis (NTA) can be considerably improved by taking into account the three-dimensional shape of the soot agglomerate including the shape factor in the calculations. 3D-TEM measurements were compared with values calculated by using a single-sphere approach when tracking nanoparticles moving under Brownian motion. A shape factor was calculated, dividing the surface area and volume calculated using spherical geometrical estimates, by the respective values calculated using the 3D models. The spherical model of the particle is found on average to overestimate the surface area by sevenfold, and the volume to the actual soot agglomerate by 23 times. Applying the calculated shape factor as a correction reduces the NTA overestimation by one order of magnitude.